In recent years mortality rates among breast cancer survivors have been declining. As a consequence, there is a growing population of breast cancer survivors. While much work in women diagnosed with breast cancer has focused on improving treatment, there has been relatively little methodologically rigorous examination of the long-term health consequences of having been diagnosed with and treated for breast cancer. This proposal builds on the Life After Cancer Epidemiology (LACE) Study cohort of breast cancer survivors that was established to examine behavioral issues related to quality of life and recurrence in the early years following diagnosis. A total of 2,364 women diagnosed with early stage breast cancer between 1996 and 2000 were enrolled in the cohort from 2000-2002. Baseline data, which was collected on average 23 months post diagnosis included information on diet, physical activity, weight, comorbid conditions and quality of life. Women are followed semiannually for recurrence and death. During the proposed new funding period, the cohort will be re-surveyed by mail to update exposures as close to their five-year post-diagnosis date as possible (mean time 6.0 yr.). This study proposes to follow these women for another four years and will focus primarily on behavioral and medical predictors of recurrence, survival, and quality of life in the permanent survival period and late and long term health effects of cancer therapy. We will also study health care utilization and costs in breast cancer survivors compared to a matched comparison group free of cancer. This study will achieve tremendous cost savings by examining the proposed questions within an existing cohort of cancer survivors for whom data on behavioral and medical factors post-diagnosis are already available. In addition, since 80 percent of the cohort belongs to Kaiser Permanente of Northern California (KPNC) Medical Care Program, self-reported outcomes can be validated and other outcomes ascertained through medical chart review and linkage to computerized databases that routinely capture health care data relevant to the proposed study aims.